This aggregates results from Ghost's three security scanners (dependencies, secrets, and code) into a single prioritized markdown report. It reads findings from each scan type, filters for high-confidence issues only (confirmed exploitable deps, verified code flaws, non-rejected secrets), then sorts by severity and inlines all the details so you get one complete document instead of hunting through individual finding files. The filtering logic is opinionated in a good way: it skips unverified dependency vulns and rejected secrets entirely, which cuts noise. Use it when you want the executive summary after running multiple scans, or when you need to hand off a security audit without explaining the toolchain. It caches reports per commit, so re-running is instant.
npx -y skills add ghostsecurity/skills --skill ghost-report --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
hoodini/ai-agents-skills
addyosmani/agent-skills
giuseppe-trisciuoglio/developer-kit
agamm/claude-code-owasp